A policy network learns to choose unmasking order in masked diffusion by reweighting the loss, outperforming random and heuristic baselines on ordering-sensitive tasks.
Accelerated sampling from masked diffusion models via entropy bounded unmasking
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
representative citing papers
SHADOWMASK backdoors MDLMs by replacing the all-mask terminal distribution with a trigger-mask mixture prior, achieving near-100% attack success on DiT and LLaDA-8B models across multiple datasets while resisting fine-tuning and some defenses.
LEAP detects early-converging tokens in dLLMs via future context filtering and multi-sequence superposition, reducing average denoising steps by about 30% while maintaining accuracy.
DMax uses On-Policy Uniform Training and Soft Parallel Decoding to enable aggressive parallelism in dLLMs, raising TPF on GSM8K from 2.04 to 5.47 and on MBPP from 2.71 to 5.86 while preserving accuracy.
Theoretical analysis reveals MaskGIT's implicit temperature sampling in masked diffusion; proposes equivalent moment sampler and efficiency techniques for adaptive unmasking with image and text experiments.
Fast-dLLM++ generalizes Fast-dLLM decoding to heterogeneous confidence profiles via Fréchet profile selection, delivering up to 37% throughput gains on GSM8K, MATH, HumanEval, and MBPP with LLaDA-8B.
PulseCol introduces periodically refreshed column-sparse attention to achieve up to 1.95x speedup over FlashAttention in diffusion LLMs with maintained model quality.
Diffusion LLMs can act as their own efficiency teachers by using revokable parallel decoding to identify reliable token orders and then distilling those orders into the model parameters for faster inference.
Stability-Weighted Decoding improves diffusion LLM accuracy by modulating token scores with temporal stability from KL divergence between prediction steps.
DLMs exhibit lower n-gram entropy, higher semantic coherence, and higher semantic diversity than ARMs, primarily due to bidirectional context and remasking decoding strategies.
citing papers explorer
-
DMax: Aggressive Parallel Decoding for dLLMs
DMax uses On-Policy Uniform Training and Soft Parallel Decoding to enable aggressive parallelism in dLLMs, raising TPF on GSM8K from 2.04 to 5.47 and on MBPP from 2.71 to 5.86 while preserving accuracy.